How to write a command on Formula-Character

bernie426

I will like to know how to write a workable command using Formula-Character functionality.

For example, I have a column of character data that have data like leveled from Class 1.0 to Class 9.0.

I would like convert this information to another column that only show 1.0 to 9.0, how can I use the formula character function to write a command for this?

The other request I like to know is how can I change a column of character data to another statement in a new column.

For example, I have a character column of data showing sweet, tasty, sour,...and I would like to change it to be sweet-->Bin1, tasty-->Bin2, sour-->Bin3. How can I use the formula character to convert the sweet-->Bin1 in a new column.

Thanks,

jules

Hi bernie426,

To extract a piece of a your column you could use the Word() function, which takes two arguments: the first is which "word" to extract (you will enter a number here), and the second is the column with text you wish to extract from. There is a third, optional argument in which you can specify the delimiters that separate words, but in your case the default of a "space" will work. So, to use this, make a new formula column with Word(2,:Class)  (where :Class is whatever variable your original data is stored in).  For example (also in the attached dataset):

This will take the second "word" which is just whatever is after the first space encountered, thus just the numbers.

For your second question I would use the match function (below). You can write this yourself, or use the column recode utility to write it for you. To do the latter, select the original column, then go to Cols > Recode. Specify what new values you wish your variable to have, then select "Formula Column" instead of "In Place."  JMP will make you a new column using the match function.

You could use Cols > Recode in this same way for your first question, but if you have many levels using Word() will be faster than manually specifying the recoding.

I've attached an example dataset with these formulae.

I hope this helps!

julian

example.jmp